Hi Michael,
I've also stumbled upon this issue. My host is Debian 9, guest is
Debian 6. This issue happens with any Qemu version from 2.4 up to 2.11
- faulty Seabios doesn't work on any of them, good Seabios works on
all of them. Guest kernel isn't even loaded - qemu gets stuck at 100%
cpu. My command line to start VM is the following:
qemu-system-x86_64 \
-enable-kvm \
-bios /opt/seabios/out/bios.bin \
-machine q35,mem-merge=on,vmport=off \
-m 2048 \
-nodefaults \
-no-hpet \
-rtc base=utc,clock=host \
-boot order=d \
-name "VM_NAME" \
-cpu host,kvm=off \
-smp 4,sockets=1,cores=4,threads=1 \
-drive
file="disk.img",id=disk,format=raw,if=none,cache=directsync,aio=native
\
-device virtio-blk-pci,drive=disk,scsi=off,config-wce=off \
-device
virtio-net-pci,netdev=net0,mac=$("../qemu-mac-hasher.py"
"VM_NAME"),mq=on,vectors=10 \
-netdev
tap,id=net0,script="../ifup.sh",downscript="../ifdown.sh",vhost=on,queues=4
\
-vga none \
-nographic
If I change '-vga none' to '-vga std', it works on all faulty Seabios
versions. I've performed bisect on Seabios, and here's the result
1d9e87b937d646be1950695f9ead35100d5ebbe6 is the first bad commit
commit 1d9e87b937d646be1950695f9ead35100d5ebbe6
Author: Gerd Hoffmann <kra...@redhat.com>
Date: Fri Jun 26 09:44:00 2015 +0200
virtio: run drivers in 32bit mode
virtio version 1.0 registers can (and actually do in the qemu
implementation) live in mmio space. So we must run the blk and
scsi virtio drivers in 32bit mode, otherwise we can't access them.
This also allows to drop a bunch of GET_LOWFLAT calls from the virtio
code in the following patches.
Signed-off-by: Gerd Hoffmann <kra...@redhat.com>
:100644 100644 f97b1bd9f91df7417e7bd5a1ebd7182098a43296
e287530d0008e1d876a0bb9f9ff8a43266d7e1bd M Makefile
:040000 040000 70b419ced4f58a39143c92897097a08b5548cbf4
93f7d4d939b5e0a14102a3cb718b1be5a2f12e60 M src
Best regards,
Sergey Aleynikov
